Embeddings
Embeddings convert text into dense vectors for semantic search and retrieval workflows.
In Lyntaris V2, embedding models are selected securely and centrally as the first configuration step for any Document Store (/vector-databases). You do not need to construct complex chains or drag embedding nodes onto a canvas.
The Lyntaris Way (Document Store Wizard)
When creating or configuring a new Document Store (such as Weaviate, Qdrant, or Pinecone), the system will prompt you to select an Embedding Provider.
Recommended Default
The platform inherently supports Local E5 Embeddings (FastApiE5Embedding), which runs locally on the Lyntaris Host GPU. This is the recommended service for enterprise workflows that process sensitive data or wish to avoid cloud-provider API costs.
Cloud Providers
If you prefer cloud-managed embeddings, Lyntaris supports all major providers. Simply leverage the Shared Credentials pipeline to authenticate securely without pasting API keys into individual configurations.